Reproduce “Trust-Region Eigenvalue Filtering for Projected Newton” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Honglin Chen, Hsueh-Ti Derek Liu, Alec Jacobson, David I.W. Levin, Changxi Zheng
The authors introduce an adaptive Hessian-filtering rule for quasistatic deformable-solid simulation. It uses the fit between a local quadratic energy model and energy decrease to choose between clamping negative eigenvalues and replacing them with their magnitudes. Across meshes, it favors adaptive filtering in large-deformation, high-Poisson’s-ratio cases and is comparable in milder cases, without establishing an elapsed-time advantage.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.